hi-RF: Incremental Learning Random Forest for large-scale multi-class Data Classification
This work addresses the problem of scalable classification for applications with evolving data and classes, but it is incremental as it builds on existing random forest and nearest class mean techniques.
The paper tackles the challenge of large-scale multi-class classification with dynamically growing data and classes by proposing hi-RF, an incremental learning random forest method that balances accuracy and computational efficiency, achieving comparable precision with significantly reduced computational time in experiments.
In recent years, dynamically growing data and incrementally growing number of classes pose new challenges to large-scale data classification research. Most traditional methods struggle to balance the precision and computational burden when data and its number of classes increased. However, some methods are with weak precision, and the others are time-consuming. In this paper, we propose an incremental learning method, namely, heterogeneous incremental Nearest Class Mean Random Forest (hi-RF), to handle this issue. It is a heterogeneous method that either replaces trees or updates trees leaves in the random forest adaptively, to reduce the computational time in comparable performance, when data of new classes arrive. Specifically, to keep the accuracy, one proportion of trees are replaced by new NCM decision trees; to reduce the computational load, the rest trees are updated their leaves probabilities only. Most of all, out-of-bag estimation and out-of-bag boosting are proposed to balance the accuracy and the computational efficiency. Fair experiments were conducted and demonstrated its comparable precision with much less computational time.